Comparison of AC and DC LED Light Bulb Efficiency for the ...
Transcript of Comparison of AC and DC LED Light Bulb Efficiency for the ...
Comparison of AC and DC LED Light Bulb
Efficiency for the DC House Project
By
Brandon Stafford
Faculty Advisor
Taufik
Senior Project
ELECTRICAL ENGINEERING DEPARTMENT
California Polytechnic State University
San Luis Obispo
2017
1
Abstract
This project investigates the efficiency of using LED light bulbs when operated under two
different electrical systems: AC and DC systems. The project is in conjunction with the
implementation of the DC House system in urban areas. More specifically, this project looks at
whether having a hybrid AC and DC system in a typical urban house will increase the efficiency
in operating the LED light bulbs, and thus reducing energy consumption and cost. The efficiency
of the LED light bulbs are measured by looking at their lumens per watt output. Several light
bulbs at four different wattage levels were tested. For the DC light bulbs, further tests were
performed at two voltage levels: 12V and 48V. Results indicate the importance of selecting the
proper converters when DC system is to be used in a hybrid AC/DC house. Results also
demonstrate the critical choice of DC voltage level when DC electrical distribution system is
planned to be installed inside the hybrid house.
2
Contents
LIST OF FIGURES .............................................................................................................................................. 3
LIST OF TABLES ................................................................................................................................................ 4
1. INTRODUCTION ................................................................................................................................... 5
2. BACKGROUND ..................................................................................................................................... 7
3. EXPERIMENT DESIGN ......................................................................................................................... 10
4. DATA AND RESULTS ........................................................................................................................... 19
5. CONCLUSION AND FURTHER WORK ................................................................................................... 26
6. REFERENCES ...................................................................................................................................... 28
APPENDIX A – ANALYSIS OF SENIOR PROJECT DESIGN .................................................................................. 30
3
List of Figures
Figure 3-1 Experimental Box ................................................................................................... 10
Figure 3-2 Top of box with opening for light sensor .............................................................. 11
Figure 3-3 Block diagram for AC light bulb test (top) and DC light bulb test (bottom) .......... 11
Figure 3-4 Experiment setup for DC input. ............................................................................. 12
Figure 4-1 Lumens per watt plot with trend lines for LED Lights with bulb ........................... 22
Figure 4-2 LED light plot with Trend lines without bulb ......................................................... 23
Figure 4-3 Lumens vs Watt for 5W and 10W LED light with bulb .......................................... 24
Figure 4-4 Lumens vs Watt for 5W and 10W LED light without bulb ..................................... 25
4
List of Tables
Table 3-1 Two light bulbs at 3W, 5W, and 10W for 120Vac, 48Vdc and 12Vdc ..................... 12
Table 3-2 A single light bulb at 3W, 5W, 7W, and 10W for 120Vac, 48Vdc and 12Vdc ......... 14
Table 3-3 Two light bulbs at 3W, 5W, 7W, and 10W at 120Vac, 48Vdc and 12Vdc ............... 15
Table 3-4 A single light bulb at 3W, 5W, and 10W for 120Vac, 48Vdc and 12Vdc................. 16
Table 3-5 A single light bulb at 3W, 5W, 7W, and 10W for 120Vac and 48Vdc ..................... 17
Table 3-6 Two Light bulbs at 3W, 5W, 7W, and 10W for 120Vac and 48Vdc ........................ 18
Table 4-1 Data with bulb attached ......................................................................................... 19
Table 4-2 Data collected for LED light bulb without bulb ....................................................... 20
Table 4-3 Average percent increase of Lumens compared to AC light bulb. ......................... 21
Table 4-4 Average percent increase of power usage compared to AC light bulb. ................. 21
5
1. Introduction
Today’s society has become so dependent on electricity that it is now taken for granted.
However, electricity is in fact not available everywhere and is not always affordable. The
International Energy Agency reports there are 1.2 billion people without electricity in the world
today. Of these 1.2 billion people, approximately 80% live in rural areas which remain
undeveloped [2]. In order to build the infrastructure required to expand the alternating current
(AC) grid and provide electricity to many of these rural areas is a costly endeavor. One solution
to this problem is the establishment of micro-grids and generate the power locally and directly
in the area. To achieve this, renewable resources such as photovoltaics, hydroelectric, and
wind power generation can be built and placed near these communities. The direct current
(DC) House project was proposed to provide a residential electrical system operating fully on
DC electricity and utilizes renewable power sources to generate electricity to provide power to
these remote communities.
The DC House, as the name implies, is a living environment that utilizes DC power to
provide the needed electricity for light, refrigeration and many other conveniences many
people in urban areas enjoy. The reason for using the DC power is mainly due to the increased
popularity of residential roof-top solar panels. These solar panels produce DC electricity, so
people who live in urban areas with solar panels on the roof of their house will have to convert
the power generated from DC to AC power by means of an inverter. However, through the use
of inverter there is an intermediate power loss that is undesirable. This power loss creates an
unnecessary load on the system resulting in a greater operational cost and increase in initial
build cost through the introduction of power inverters and the requirement of additional
6
generators to make up the loss. This problem exacerbates when the load is DC such as LED
lights, portable electronics, etc. Therefore, the use of DC electrical system allows the
elimination of the power inverters from the solar panels which makes it more energy efficient.
The DC house provides great benefits for those who live out in the rural areas by reducing
the cost to provide power and allows the creation of micro-grids for these communities. By
utilizing a DC bus we can minimize the reactive power loss of low power devices such as LED
light bulbs and various chargers by eliminating the need of AC to DC converter for each one.
For those living in urban areas, where the power systems can be tied into the AC power grid, a
simpler solution may be the use of a single AC to DC converter to provide a DC bus to provide
electricity to low power devices. This leaves the higher power devices such as washing
machines, dryers, and refrigerators, devices that typically have a high reactive load, powered by
AC.
7
2. Background
The battle between AC and DC started with Edison and Tesla. The invention of the
transformer enabled the transmission of power over long distances with keep down power loss
on the transmission lines [4]. This allowed AC to win out since it reduced the number of power
generators and increased the distance from the source that power can be sent. With current
technology, namely power electronics, we can generate DC electricity locally and step up or
down the voltage to make it usable for homes. This means that technology can be developed to
run off of DC power and make modern conveniences available to people living in rural areas
where it is too expensive to build the infrastructure for AC power to be generated and
transmitted there. Because of this, the DC House project was formed in a humanitarian effort
to bring low cost power to rural communities that have done without so far [3,10].
The first phase of the DC house project was started in September 2010 and completed in
June 2011. This phase involved the development of designs for a Multi-input DC-DC converter,
a variable voltage DC wall outlet, and power generation with Photo-voltaic, Wind, hydro-
electric, and bicycle power systems. The Multi-Input Single-Output (MISO) converter is one of
the key projects that was completed during this phase. The MISO DC-DC Converter allows wind,
solar, hydro, bicycle, and future DC power generators to be connected to a single DC bus for the
use by systems in the house or to charge a battery system [5].
The second phase of the DC house project, active from 2011 to 2012, includes the
improvement of the MISO converter and DC wall outlet; as well as development of a DC light
bulb, cell phone charger, and a Seesaw Human Powered Generator. The DC LED light bulb was
developed and incorporated an LED driver that allowed the light bulb to function off of inputs
8
from 24VDC to 72VDC and provides a highly efficient means of producing light. The use of LED
Light bulbs as the primary light source provides highly efficient low power means of lighting the
DC House [6].
The third phase of the DC house project, active from 2012 to 2013, includes the
development of a DC distribution panel, portable DC light bulb, Swing Human-powered
generator, Merry-Go-Round Human powered generator, and bi-directional DC-DC flyback
converter. The bidirectional flyback converter enables the inclusion of a battery backup system
to the DC house. Additionally, the power distribution system ensures that a proper power flow
is maintained [7].
The fourth phase of the DC house project, active from 2014 to 2015, includes the
development of an Arc Fault Current Interrupter (AFCI) and ventilation wind power generator,
as well as improvements to the Merry-Go-Round human-powered generator, Swing human-
powered generator, portable Nano hydro generator, Bi-directional DC-DC flyback converter and
the DC Light bulb. The AFCI is an important addition to the DC House that improves safety as it
detects when arcing occurs and responds appropriately to eliminate the arc [8].
The fifth phase of the DC House project started in September 2015 and is ongoing. The
goals for this phase is the development of 3 fully functioning DC House prototypes,
Development of improved swing and merry-go-round generators, and Development of an
ocean-wave portable generator. The prototypes are located at Cal Poly, Universitas
Padjadjaran Bandung Indonesia, and Technological Institute of the Philippines [9].
This project is part of the DC House project to investigate the efficiency of low power
devices when comparing their use between AC and DC. By looking at the power usage for a
9
device on AC and then the power usage of a DC equivalent device with a high efficiency Switch
Mode Power Supply (SMPS) and comparing them. Based on the results, we can determine if
using a DC light bulb or other low powered device run off of a bus powered by an SMPS is more
energy efficient than running their AC counterpart. From this we can determine if a DC bus can
be used to complement the existing AC system for houses in urban areas that are tied into the
existing AC grid to keep electricity cost down.
10
3. Experiment Design
To compare the efficiency of low power devices between AC and DC counterparts, we
decided to compare LED light bulbs for the purposes of the DC house in urban areas. Since light
bulbs are rated by their wattage and not by their lumen output, although many have a lumen
rating listed, we decided to compare the light bulbs by looking at their lumens per watt output.
Since the high reactive load components will be running off of AC and the DC will be provided
by a SMPS powered by AC we would look at the power used by both the SMPS and the DC light
bulb vs the AC light bulb by itself.
In order to provide a consistent measurement and minimize the impact of external light
sources on the measurements for light output, a box was constructed to house the standard
A19 socket and block out external light sources, shown in figure 3-1.
Figure 3-1 Experimental Box
The box was constructed out of scrap wood to keep cost down. The A19 socket was wired
to 2 female banana plugs to allow for use with both AC and DC power sources. A hole was
11
drilled into the top of the box, as shown in Figure 3-2, to provide an opening to mount the light
sensor of the lux meter.
Figure 3-2 Top of box with opening for light sensor
The opening was sized to the light sensor aperture on the lux meter to minimize the amount
of external light entering the box and allow for more accurate measurements. The block
diagram in Figure 3-3 shows the experiment design as collected.
AC Source Watt Meter AC Lightbulb
AC Source Watt MeterAC/DC
ConverterDC Lightbulb
Figure 3-3 Block diagram for AC light bulb test (top) and DC light bulb test (bottom)
Figure 3-4 shows the laboratory setup for the data collection of the DC light bulbs.
12
Figure 3-4 Experiment setup for DC input.
To provide sufficient data, light bulbs with varying wattage voltage levels at 12V and 48V
were selected. There were 4 options for light bulb wattage for the DC light bulb: 3W, 5W, 7W,
and 10W. Again to minimize cost, 3 power levels were chosen: the 3W, 5W, and 10W, and their
AC equivalents. The SMPS power supply for the DC light bulbs was chosen as a 25W 90%
efficient AC/DC converter that outputs 48V and a second one that outputs 12V. Tables 3-1 to 3-
6 show a cost assessment of each design that was considered. The design chosen was Table 3-
1, having more than a single light bulb to test provides additional data as the lumens output
and power usage of each light can vary.
Table 3-1 Two light bulbs at 3W, 5W, and 10W for 120Vac, 48Vdc and 12Vdc
13
Qty Cost/unit Total Cost
3W 48Vdc Light bulb 2 $16.85 $33.70
3W 12Vdc Light bulb 2 $15.85 $31.70
3W AC Light bulb (6 count) 1 $16.99 $16.99
5W 48Vdc Light bulb 2 $20.25 $40.50
5W 12Vdc Light bulb 2 $19.65 $39.30
5W AC Light bulb (2 count) 1 $14.99 $14.99
10W 48Vdc Light bulb 2 $25.65 $51.30
10W 12Vdc Light bulb 2 $22.50 $45.00
10W AC Light bulb (3 Count) 1 $14.95 $14.95
Pass & Seymour White Medium Base Cleat Lamp Socket 1 $2.99 $2.99
Lux Meter 1 $18.99 $18.99
AC/DC Converter 48V (25W) 1 $25.00 $25.00
AC/DC Converter 12V (25W) 1 $25.00 $25.00
Molex 09-50-10-31 2 $0.45 $0.90
Molex 09-50-10-41 2 $0.26 $0.52
SPOX 5194 14 $0.10 $1.40
Hookup Wire 20AWG 1 $8.99 $8.99
Total: $372.22
Incidentally, the ordered DC Light bulbs received were 12V-80V DC light bulbs which
allowed for more data to be collected for each of the DC cases, but may have an impact on the
bulbs LED efficiency.
14
Table 3-2 A single light bulb at 3W, 5W, 7W, and 10W for 120Vac, 48Vdc and 12Vdc
Qty Cost/unit Total Cost
3W 48Vdc Light bulb 1 $16.85 $16.85
3W 12Vdc Light bulb 1 $15.85 $15.85
3W AC Light bulb (6 count) 1 $16.99 $16.99
5W 48Vdc Light bulb 1 $20.25 $20.25
5W 12Vdc Light bulb 1 $19.65 $19.65
5W AC Light bulb (2 count) 1 $14.99 $14.99
7W 48Vdc Light bulb 1 $23.85 $23.85
7W 12Vdc Light bulb 1 $21.80 $21.80
7W AC Light bulb (2 count) 1 $19.99 $19.99
10W 48Vdc Light bulb 1 $25.65 $25.65
10W 12Vdc Light bulb 1 $22.50 $22.50
10W AC Light bulb (3 Count) 1 $14.95 $14.95
Pass & Seymour White Medium Base Cleat Lamp Socket 1 $2.99 $2.99
Lux Meter 1 $18.99 $18.99
AC/DC Converter 48V (25W) 1 $25.00 $25.00
AC/DC Converter 12V (25W) 1 $25.00 $25.00
Molex 09-50-10-31 2 $0.45 $0.90
Molex 09-50-10-41 2 $0.26 $0.52
SPOX 5194 14 $0.10 $1.40
Hookup Wire 20AWG 1 $8.99 $8.99
Total: $317.11
15
Table 3-3 Two light bulbs at 3W, 5W, 7W, and 10W at 120Vac, 48Vdc and 12Vdc
Qty Cost/unit Cost
3W 48Vdc Light bulb 2 $16.85 $33.70
3W 12Vdc Light bulb 2 $15.85 $31.70
3W AC Light bulb (6 count) 1 $16.99 $16.99
5W 48Vdc Light bulb 2 $20.25 $40.50
5W 12Vdc Light bulb 2 $19.65 $39.30
5W AC Light bulb (2 count) 1 $14.99 $14.99
7W 48Vdc Light bulb 2 $23.85 $47.70
7W 12Vdc Light bulb 2 $21.80 $43.60
7W AC Light bulb (2 count) 1 $19.99 $19.99
10W 48Vdc Light bulb 2 $25.65 $51.30
10W 12Vdc Light bulb 2 $22.50 $45.00
10W AC Light bulb (3 Count) 1 $14.95 $14.95
Pass & Seymour White Medium Base Cleat Lamp Socket 1 $2.99 $2.99
Lux Meter 1 $18.99 $18.99
AC/DC Converter 48V (25W) 1 $25.00 $25.00
AC/DC Converter 12V (25W) 1 $25.00 $25.00
Molex 09-50-10-31 2 $0.45 $0.90
Molex 09-50-10-41 2 $0.26 $0.52
SPOX 5194 14 $0.10 $1.40
Hookup Wire 20AWG 1 $8.99 $8.99
Total: $483.51
16
Table 3-4 A single light bulb at 3W, 5W, and 10W for 120Vac, 48Vdc and 12Vdc
Qty Cost/unit Total Cost
3W 48Vdc Light bulb 1 $16.85 $16.85
3W 12Vdc Light bulb 1 $15.85 $15.85
3W AC Light bulb (6 count) 1 $16.99 $16.99
5W 48Vdc Light bulb 1 $20.25 $20.25
5W 12Vdc Light bulb 1 $19.65 $19.65
5W AC Light bulb (2 count) 1 $14.99 $14.99
10W 48Vdc Light bulb 1 $25.65 $25.65
10W 12Vdc Light bulb 1 $22.50 $22.50
10W AC Light bulb (3 Count) 1 $14.95 $14.95
Pass & Seymour White Medium Base Cleat Lamp Socket 1 $2.99 $2.99
Lux Meter 1 $18.99 $18.99
AC/DC Converter 48V (25W) 1 $25.00 $25.00
AC/DC Converter 12V (25W) 1 $25.00 $25.00
Molex 09-50-10-31 2 $0.45 $0.90
Molex 09-50-10-41 2 $0.26 $0.52
SPOX 5194 14 $0.10 $1.40
Hookup Wire 20AWG 1 $8.99 $8.99
Total: $251.47
17
Table 3-5 A single light bulb at 3W, 5W, 7W, and 10W for 120Vac and 48Vdc
Qty Cost/unit Total Cost
3W 48Vdc Light bulb 1 $16.85 $16.85
3W AC Light bulb (6 count) 1 $16.99 $16.99
5W 48Vdc Light bulb 1 $20.25 $20.25
5W AC Light bulb (2 count) 1 $14.99 $14.99
7W 48Vdc Light bulb 1 $23.85 $23.85
7W AC Light bulb (2 count) 1 $19.99 $19.99
10W 48Vdc Light bulb 1 $25.65 $25.65
10W AC Light bulb (3 Count) 1 $14.95 $14.95
Pass & Seymour White Medium Base Cleat Lamp Socket 1 $2.99 $2.99
Lux Meter 1 $18.99 $18.99
AC/DC Converter 48V (25W) 1 $25.00 $25.00
Molex 09-50-10-31 2 $0.45 $0.90
Molex 09-50-10-41 2 $0.26 $0.52
SPOX 5194 14 $0.10 $1.40
Hookup Wire 20AWG 1 $8.99 $8.99
Total: $212.31
18
Table 3-6 Two Light bulbs at 3W, 5W, 7W, and 10W for 120Vac and 48Vdc
Qty Cost/unit Total Cost
3W 48Vdc Light bulb 2 $16.85 $33.70
3W AC Light bulb (6 count) 1 $16.99 $16.99
5W 48Vdc Light bulb 2 $20.25 $40.50
5W AC Light bulb (2 count) 1 $14.99 $14.99
7W 48Vdc Light bulb 2 $23.85 $47.70
7W AC Light bulb (2 count) 1 $19.99 $19.99
10W 48Vdc Light bulb 2 $25.65 $51.30
10W AC Light bulb (3 Count) 1 $14.95 $14.95
Pass & Seymour White Medium Base Cleat Lamp Socket 1 $2.99 $2.99
Lux Meter 1 $18.99 $18.99
AC/DC Converter 48V (25W) 1 $25.00 $25.00
Molex 09-50-10-31 2 $0.45 $0.90
Molex 09-50-10-41 2 $0.26 $0.52
SPOX 5194 14 $0.10 $1.40
Hookup Wire 20AWG 1 $8.99 $8.99
Total: $298.91
19
4. Data and Results
During the experiment, as many as 4 measurements were made for each bulb. Power in,
Lux, color temperature, and Distance from the LEDs to the lux sensor is shown in Table 4-1.
Table 4-1 Data with bulb attached
Bulb Type Pin (W) Distance (cm) Color Temperature (K) Lux Lumens
10W 120Vrms 9 12.4 2890K 9340 902.340
10W 120Vrms 8.96 12.4 2884K 9960 962.238
10W 120Vrms 9 12.4 2886K 10650 1028.899
5W 120Vrms 4.37 13.3 3008K 4140 460.133
5W 120Vrms 4.48 13.3 3068K 4170 463.467
5W 120Vrms 4.24 13.3 3049K 3960 440.127
3W 120Vrms 4.28 13.9 2799K 2030 246.437
3W 120Vrms 4.37 13.9 2793K 2040 247.651
3W 120Vrms 4.25 13.9 2795K 1990 241.581
10W 48 Vdc 11.38 10.7 2942K 15200 1093.430
10W 48 Vdc 10.61 10.7 2942K 14900 1071.849
10W 48 Vdc 11.08 10.7 2925K 15300 1100.624
5W 48 Vdc 7.13 13.2 2990K 5540 606.509
5W 48 Vdc 7.02 13.2 2978K 5550 607.604
5W 48 Vdc 7.07 13.2 2970K 5600 613.078
3W 48 Vdc 4.6 14.1 3021K 2190 273.566
3W 48 Vdc 4.62 14.1 3085K 2190 273.566
3W 48 Vdc 4.41 14.1 3061K 2130 266.071
10W 12 Vdc 11.81 10.7 2930K 15600 1122.205
10W 12 Vdc 11.04 10.7 2932K 14800 1064.656
10W 12 Vdc 11.4 10.7 2935K 15100 1086.236
5W 12 Vdc 7.27 13.2 2985K 5900 645.922
5W 12 Vdc 7 13.2 2957K 5659 619.537
5W 12 Vdc 7.33 13.2 2973K 5760 630.595
3W 12 Vdc 4.73 14.1 3022K 2360 294.802
3W 12 Vdc 4.6 14.1 3096K 2280 284.808
3W 12 Vdc 4.34 14.1 3076K 2150 268.569
20
The same measurements were then made again with the plastic bulb removed and the LEDs
exposed, this data is shown in Table 4-2.
Table 4-2 Data collected for LED light bulb without bulb
Bulb Type Pin (W) Distance (cm)
Color Temperature Lux Lumens
10W 120Vrms 9 12.4 2999K 17200 1661.696
10W 120Vrms 8.96 12.4 2966K 18440 1781.493
10W 120Vrms 9 12.4 2982K 18660 1802.747
5W 120Vrms 4.37 13.3 3135K 7260 806.900
5W 120Vrms 4.48 13.3 3136K 7400 822.460
5W 120Vrms 4.24 13.3 3120K 7050 783.560
3W 120Vrms 4.28 13.9 2870K 4220 512.297
3W 120Vrms 4.37 13.9 2820K 4450 540.218
3W 120Vrms 4.25 13.9 2850K 4340 526.865
10W 48 Vdc 11.38 10.7 3023K 24000 1726.468
10W 48 Vdc 10.61 10.7 3019K 23200 1668.919
10W 48 Vdc 11.08 10.7 3013K 23800 1712.081
5W 48 Vdc 7.13 13.2 3082K 9330 1021.431
5W 48 Vdc 7.02 13.2 3106K 9350 1023.621
5W 48 Vdc 7.07 13.2 3083K 9300 1018.147
3W 48 Vdc 4.6 14.1 3096K 5060 632.075
3W 48 Vdc 4.62 14.1 3194K 5040 629.576
3W 48 Vdc 4.41 14.1 3152k 4940 617.085
10W 12 Vdc 11.81 10.7 3023K 24400 1755.243
10W 12 Vdc 11.04 10.7 3030K 23400 1683.307
10W 12 Vdc 11.4 10.7 3032K 23500 1690.500
5W 12 Vdc 7.27 13.2 3102K 9660 1057.559
5W 12 Vdc 7 13.2 3100K 9460 1035.664
5W 12 Vdc 7.33 13.2 3100K 9610 1052.086
3W 12 Vdc 4.73 14.1 3102K 5360 669.549
3W 12 Vdc 4.6 14.1 3193K 5250 655.809
3W 12 Vdc 4.34 14.1 3165K 5330 665.802
Based on the limited fluctuation in color temperature, it was concluded that it will have
limited impact on the lux readings.
21
Since Lux is a measurement of Lumens per square meter, to calculate the Lumens from the
Lux measurement in Table 4-1 and Table 4-2 equations 4-1 was used.
𝐸𝑞 4 − 1 𝑀 = 𝜑 ∗ 𝐴
where φ is Luminous flux in lumens, M is luminous emittance in lux, and A is surface area in
square meters.
To calculate the surface area for the lux measurement, it was assumed that the light
projected out in front of the bulb equally distributed in a half-sphere. The equation 4-2 is used
to calculate the surface area of a half sphere. And equation 4-3 is equations 4-1 and 4-2
combined and used to calculate the Lumens of each bulb.
𝐸𝑞 4 − 2 𝐴 =4∗ 𝜋∗𝑟2
2
𝐸𝑞 4 − 3 𝜑 =2𝑀
(4 ∗ 𝜋 ∗ 𝑟2)
Table 4-3 Average percent increase of Lumens compared to AC light bulb.
10W (%) 5W (%) 3W (%)
48VDC 12.871 33.985 10.539
12VDC 13.120 39.035 15.294
Table 4-4 Average percent increase of power usage compared to AC light bulb.
10W (%) 5W (%) 3W (%)
48VDC 22.663 62.108 5.659
12VDC 27.040 65.011 5.969
𝐸𝑞 4 − 4 %𝑑𝑖𝑓𝑓 =𝐷𝐶𝑎𝑣𝑔 − 𝐴𝐶𝑎𝑣𝑔
𝐴𝐶𝑎𝑣𝑔
Equation 4-4 was used to determine the increase in lumens output and power usage for the
DC light bulb relative to the AC light bulb.
22
Lumens vs. Watts was then plotted trend lines added as shown in Figures 4-1 and 4-2.
Figure 4-1 Lumens per watt plot with trend lines for LED Lights with bulb
Figure 4-1 on the last page, shows a linear increase in the Lumens/Watt of the DC light
bulbs, but not the AC light bulbs. The 12VDC Light bulbs produces slightly more Lumens/Watt
than the 48VDC at 3W, but at 10W the 48VDC produces more lumens/watt. At 3W the AC and
DC light bulbs are roughly equivalent, but the AC light bulbs outperform the DC light bulbs at
5W and 10W.
y = 132.43x - 224.36
y = 125.26x - 289.02
y = 116.72x - 232.99
0
200
400
600
800
1000
1200
1400
0 2 4 6 8 10 12 14
Lum
ens
Watt
Lumens per Watt for LED Light with bulb
AC
48V DC
12V DC
Linear (AC)
Linear (48V DC)
Linear (12V DC)
23
Figure 4-2 LED light plot with Trend lines without bulb
Figure 4-2 shows the same results as figure 4-1. The difference between Figure 4-1 and 4-2
is the Lumens output. By removing the plastic dome from the LED light bulbs the Lumens/Watt
increases, this increase slightly intensifies the difference of the Lumens/Watt for the entire plot.
Due to the similarity in power usage of the AC 3W and 5W bulbs, Figures 4-3 and 4-4 were
plotted omitting the 3W data points to provide a better fitting trend line for the data provided.
y = 233.19x - 345.46
y = 166.17x - 137.41
y = 152.39x - 36.531
0
200
400
600
800
1000
1200
1400
1600
1800
2000
0 2 4 6 8 10 12 14
Lum
ens
Watt
Lumens per Watt for LED light Without bulb
AC
DC 48V
12V DC
Linear (AC)
Linear (DC 48V)
Linear (12V DC)
24
Figure 4-3 Lumens vs Watt for 5W and 10W LED light with bulb
Figure 4-3 shows that DC light bulbs produce more Lumens, but use more Watts to achieve
it. However, the AC light bulbs produce more Lumens/Watt than the DC. Based on the slope of
the trend lines in figure 4-3 the DC will at some point produce more Lumens/Watt than the AC
bulbs with the 48VDC light bulbs achieving that crossing before the 12VDC light bulbs.
y = 110.28x - 26.591
y = 120.23x - 239
y = 108.36x - 147.11
0
200
400
600
800
1000
1200
0 2 4 6 8 10 12 14
Lum
ens
Watt
Lumens per Watt for LED Light with bulb
AC
48V DC
12V DC
Linear (AC)
Linear (48V DC)
Linear (12V DC)
25
Figure 4-4 Lumens vs Watt for 5W and 10W LED light without bulb
Figure 4-4 shows that 5W AC bulbs produce fewer lumens at lower watts than the 5W DC
bulb, but the 10W AC produces the same amount of lumens at a lower wattage than the 10W
DC bulb. The slope of the trend line on figure 4-4 shows that the AC light bulb will maintain its
higher lumens per watt than the DC light bulbs.
y = 204.17x - 86.357
y = 171.24x - 187.66
y = 155.91x - 72.245
0
200
400
600
800
1000
1200
1400
1600
1800
2000
0 2 4 6 8 10 12 14
Lum
ens
Watt
Lumens per Watt for LED light Without bulb
AC
DC 48V
12V DC
Linear (AC)
Linear (DC 48V)
Linear (12V DC)
26
5. Conclusion and Further Work
With current light bulb technology, to run an AC LED light bulb straight off of an AC system
is more energy efficient than running a DC light bulb off of a high efficiency AC to DC SMPS. The
DC light bulbs in general produced more Lumens at their respective ratings with an average
increase of 19.1% at 48VDC and 22.5% at 12VDC. However, the DC light bulbs used an average of
30.1% more power at 48VDC and 32.7% more power at 12VDC. Though the DC Light bulbs
produce more lumens, the energy cost to do so outweighs the increase making the AC bulbs
more energy efficient.
The similar power usage of the 3W bulb and 5W bulbs for AC skews the trend line making it
difficult to extrapolate if there is a point where the DC will perform better than the AC bulb for
loading. However, by omitting the 3W bulb from the data as shown in Figures 4-3 and 4-4, a
better fitting trend line can be seen. Using the 5W and 10W bulbs only, and using Figure 4-3 it
appears that the DC light bulbs will prove more efficient as you increase the number of bulbs
but Figure 4-4 contradicts this. Comparing Figure 4-3 and 4-4 suggests that a more efficient
SMPS, such as one rated at 95% efficiency, is enough to make the DC light bulbs with an SMPS a
stronger answer as this would increase the slope for the DC light bulbs in each of the graphs
shown in Figures 4-1 through 4-4.
For the 12V DC compared the 48V DC bus, Figures 4-1 through 4-4 show that the 48V bus is
more efficient than the 12V bus when using a 10W bulb and equivalent when using a 5W bulb.
Figures 4-1 and 4-2 show that for a 3W bulb a 12V DC bus was more efficient than on the 48V
bus. This indicates that as the power draw increases the 48V DC bus is more efficient than the
12V DC bus.
27
There are two ways to expand this data for more conclusive results. The first is by
looking at the impact of the SMPS efficiency on the system and, if possible, determine an
optimal efficiency for the SMPS used. The second is by looking at the impact of multiple light
bulbs and determine the minimum number of LED light bulbs to use before the DC system
becomes more efficient than the AC system.
28
6. References
[1]. Taufik, NFN. "The DC House Project." The DC House Project. N.p., n.d. Web.
November 2016.
[2]. "Energy Poverty." IEA. International Energy Agency, n.d. Web. November 2016.
http://www.iea.org/topics/energypoverty/
[3]. Taufik. "The DC House Project: Providing Access to Electricity for the Less
Fortunate." EEWEB Pulse 10 (2011): 8-9. EEWeb. 6 Sept. 2011. Web. 29 Nov. 2016.
[4]. G. Hollings, "ABB conversations > A brief history of the power transformer," 2013.
[Online]. Available: http://www.abb-conversations.com/2013/09/a-brief-history-of-
the-power-transformer/. Accessed: Dec. 19, 2016.
[5]. M. Taufik, T. Taufik, and T. Wong, "Multiple-input single-output converter for
renewable energy sources," 2012 IEEE Symposium on Industrial Electronics and
Applications, Sep. 2012.
[6]. Liang, K., “Design of DC Light Bulb for the DC House Project”, M.S. thesis, Dept. Elect.
Eng., Cal Poly, San Luis Obispo, CA, 2012
[7]. Luan, Austin J., “Bi-Directional Flyback DC-DC Converter for Battery System of the DC
House Project”, M.S. thesis, Dept. Elect. Eng., Cal Poly, San Luis Obispo, CA, 2013
[8]. Aarstad, Cassidy A., “Arc Fault Circuit Interrupter Development for Residential DC
Electricity”, M.S. thesis, Dept. Elect. Eng., Cal Poly San Luis Obispo, CA, 2016
[9]. Taufik, F. Milan, and M. Taufik, "WIP: International Partnership on the DC House
Project for Rural Electrification", ASEE PSW Conference, April 2016.
29
[10]. Taufik, "The DC house project: An alternate solution for rural electrification," IEEE
Global Humanitarian Technology Conference (GHTC 2014), Oct. 2014.
30
Appendix A – Analysis of Senior Project Design
Project Title: Comparison of AC and DC LED Light Bulb Efficiency for the DC House Project
Student’s Name: Brandon Stafford Student’s Signature:
Advisor’s Name: Taufik Advisor’s Initials: T Date: 1/10/17
Summary of Functional Requirements
This project provides a means of collecting energy usage and lux output for LED
Lightbulbs for both AC and DC Light bulbs.
Primary Constraints
Since light bulbs are rated based on their power usage and not the lumens output, I
needed to design a system that could provide a meaningful relationship of the efficiency of LED
light bulbs. Simply hooking up a series of LED bulbs and looking at how the number of bulbs
impacted the power usage of the system would not provide meaningful data as this would
characterize the switch mode power supply instead of providing data on if using DC LED light
bulbs is more energy efficient than AC light bulbs. The next requirement was to provide a
system that would allow for reliable collection of data on the light output. Isolating the light
bulb from the rest of the room so the only light source was the light bulb eliminated the
influence of daylight or from other light sources in the lab.
Economic
The project demonstrates that when powering a light bulb off of a grid tied AC line that
using AC lightbulbs is more efficient. Incorporating a switch mode power supply increases the
31
upfront cost with no benefits to saving energy for use of lightbulbs. The benefits show at the
end of the project life cycle when it determines if DC light bulbs are more efficient than AC light
bulbs. The cost was covered by the project participant. Table 3-1 shows the initial cost, final
cost and the bill of materials.
Commercial Basis
This project has no known target market and is not expected to be produced again as
there are superiors devices that can be used to provide the same data. The price of such
existing devices is not in the budget for use of this project. To produce this device, it is
estimated to be approximately $83. This omits the cost of the light bulbs that are used for
testing.
Environmental
This project has minimal environmental impact as it is for use with existing
infrastructure. This setup makes use power generation that already exists for the AC grid. The
outcome of this project is to determine a means of minimizing power usage to provide lighting
or other low energy devices and thus reduce the amount of fossil fuels used for power
generation such as coal and natural gas. By minimizing the use of fossil fuels for energy
production, we can minimize the harm to the environment that burning these fuels has.
Manufacturability
There are few limitations on the manufacturability for this project. The components
used are readily available from manufacturers and easy to obtain. Aside from the switch mode
power supply, the design is simple, and since the SMPS is purchased and not built, it is easy to
obtain.
32
Sustainability
The maintenance of the device will be relatively simple as the primary component that
will fail is the SMPS and it is easy to switch out and it can be recycled or stripped of working
parts. The only improvement for this system would be a more efficient SMPS
Ethical
Since this project is a comparison of AC LED light bulbs and DC LED light bulbs, if this
project were to go into mass production for use by the end user it would result in an increased
cost in both electricity and materials used. As the project was designed, there is next to no
ability to misuse the project and manufacturing has no ethical problems beyond the existing
health risk of using the tools necessary to build the apparatus.
Health and Safety
The primary concerns for health and safety for the manufacture and use of this project
is the potential of cutting or crushing from use of power tools to shape the enclosure and
burning from soldering. As with any offline device, it also presents a potential fire and
electrocution risk from offline power. The fire and electrocution risk is mitigated by the use of
insulated wires.
Social and Political
This project impacts any end user who utilizes LED light bulbs. By determining if AC LED
light bulbs or DC LED light bulbs with an SMPS will effect manufacturers of both kinds of LED
light bulbs, as well as power companies and the manufacturers of SMPSs. By minimizing the
power used reduces the amount of power required impact the profit margin of power
33
companies as well as affecting costs and production of AC LED light bulbs and DC LED light
bulbs.
Development
While developing this project, I looked into how light was measured and how to use
these measurements to determine the efficiency of light bulbs, in particular LED light bulbs, in
their conversion of watts to lumens. This required learning how to calculate the approximate
lumens output based on a lux reading.